Showing papers by "Xihong Lin published in 2006"

Semiparametric estimation in general repeated measures problems

Abstract: Summary. The paper considers a wide class of semiparametric problems with a parametric part for some covariate effects and repeated evaluations of a nonparametric function. Special cases in our approach include marginal models for longitudinal or clustered data, conditional logistic regression for matched case–control studies, multivariate measurement error models, generalized linear mixed models with a semiparametric component, and many others. We propose profile kernel and backfitting estimation methods for these problems, derive their asymptotic distributions and show that in likelihood problems the methods are semiparametric efficient. Although generally not true, it transpires that with our methods profiling and backfitting are asymptotically equivalent. We also consider pseudolikelihood methods where some nuisance parameters are estimated from a different algorithm. The methods proposed are evaluated by using simulation studies and applied to the Kenya haemoglobin data.

Impact of Physician Asthma Care Education on Patient Outcomes

Abstract: OBJECTIVE. We evaluated the effectiveness of a continuing medical education program, Physician Asthma Care Education, in improving pediatricians' asthma therapeutic and communication skills and patients' health care utilization for asthma. METHODS. We conducted a randomized trial in 10 regions in the United States. Primary care providers were recruited and randomly assigned by site to receive the program provided by local faculty. The program included 2 interactive seminar sessions (2.5 hours each) that reviewed national asthma guidelines, communication skills, and key educational messages. Format included short lectures, case discussions, and a video modeling communication techniques. We collected information on parent perceptions of physicians' communication, the child's asthma symptoms, and patients' asthma health care utilization. We used multivariate regression models to determine differences between control and intervention groups. RESULTS. A total of 101 primary care providers and a random sample of 870 of their asthma patients participated. After 1 year, we completed follow-up telephone interviews with the parents of 731 of the 870 patients. Compared to control subjects, parents reported that physicians in the intervention group were more likely to inquire about patients' concerns about asthma, encourage patients to be physically active, and set goals for successful treatment. Patients of physicians that attended the program had a greater decrease in days limited by asthma symptoms (8.5 vs 15.6 days), as well as decreased emergency department asthma visits (0.30 vs 0.55 visits per year). CONCLUSIONS. The Physician Asthma Care Education program was used in a range of locations and was effective in improving parent-reported provider communication skills, the number of days affected by asthma symptoms, and asthma health care use. Patients with more frequent asthma symptoms and higher health care utilization at baseline were more likely to benefit from their physician's participation in the program.

Impact of physician asthma care education on patient outcomes. - eScholarship

Abstract: OBJECTIVE. We evaluated the effectiveness of a continuing medical education program, Physician Asthma Care Education, in improving pediatricians' asthma therapeutic and communication skills and patients' health care utilization for asthma. METHODS. We conducted a randomized trial in 10 regions in the United States. Primary care providers were recruited and randomly assigned by site to receive the program provided by local faculty. The program included 2 interactive seminar sessions (2.5 hours each) that reviewed national asthma guidelines, communication skills, and key educational messages. Format included short lectures, case discussions, and a video modeling communication techniques. We collected information on parent perceptions of physicians' communication, the child's asthma symptoms, and patients' asthma health care utilization. We used multivariate regression models to determine differences between control and intervention groups. RESULTS. A total of 101 primary care providers and a random sample of 870 of their asthma patients participated. After 1 year, we completed follow-up telephone interviews with the parents of 731 of the 870 patients. Compared to control subjects, parents reported that physicians in the intervention group were more likely to inquire about patients' concerns about asthma, encourage patients to be physically active, and set goals for successful treatment. Patients of physicians that attended the program had a greater decrease in days limited by asthma symptoms (8.5 vs 15.6 days), as well as decreased emergency department asthma visits (0.30 vs 0.55 visits per year). CONCLUSIONS. The Physician Asthma Care Education program was used in a range of locations and was effective in improving parent-reported provider communication skills, the number of days affected by asthma symptoms, and asthma health care use. Patients with more frequent asthma symptoms and higher health care utilization at baseline were more likely to benefit from their physician's participation in the program.

Genotypes and haplotypes of matrix metalloproteinase 1, 3 and 12 genes and the risk of lung cancer

Abstract: The MMPs (matrix metalloproteinases) are a family of secreted zinc metalloproteases that degrade the collagens of the extracellular matrix important in tissue remodeling and repair during development and inflammation. We investigated the associations between polymorphisms of MMP-1 (-1607 1G/2G, rs1799750), MMP-3 (-1171 5A/6A, rs3025058), and MMP-12 (-82AG, rs2276109, and 1082A/G, rs652438) and the risk of lung cancer in 2014 Caucasian lung cancer patients and 1323 healthy controls. The results were analyzed using logistic regression models, adjusting for covariates. The four polymorphisms were in Hardy-Weinberg disequilibrium. Except for the 1G-1082A, the other linkage disequilibrium tests between the four MMP polymorphisms were statistically significant (P < 0.001). There was no overall association between individual MMP polymorphism and the risk of lung cancer. The MMP polymorphisms jointly were associated with a non-statistically significant higher risk of lung cancer, with the adjusted odds ratio (AOR) of subjects with 5+ variant alleles versus zero variant allele of 1.31 [95% confidence interval (CI), 0.92-1.88]. Stronger associations were observed in never-smokers and males, with the corresponding AORs of 2.44 (95%CI, 1.10-5.43, P(trend) = 0.04) in never smokers and 1.35 (95%CI, 0.79-2.30, P(trend) = 0.04) in men. In haplotype analysis, the 1G-6A-82A-1082G haplotype was associated with higher risk of lung cancer among never smokers, with the AOR of 3.65 (95%CI, 1.62-8.20) when compared with the most common 1G-5A-82A-1082A haplotype. In conclusion, the combined MMP genotypes and associated haplotypes may be associated with higher risk of lung cancer, particularly among never smokers and men.

Recent Exposure to Particulate Matter and C-reactive Protein Concentration in the Multi-Ethnic Study of Atherosclerosis

Abstract: for person-level covariates were 3% (95% confidence interval (CI): –2, 10) for a 30-day mean and 4% (95% CI: –3, 11.0) for a 60-day mean. The means of 7-day, 30-day, and 60-day exposures were weakly, positively, and nonsignificantly associated with the odds of C-reactive protein of greater than or equal to 3 mg/liter: adjusted odds ratios were 1.05 (95% CI: 0.96, 1.15), 1.12 (95% CI: 0.98, 1.29), and 1.12 (95% CI: 0.96, 1.32), respectively. Slightly stronger associations were observed in persons without other risk factors for elevated C-reactive protein, but this heterogeneity was not statistically significant. The authors’ results are not compatible with strong effects of particulate matter exposures on population levels of C-reactive protein. air pollutants, environmental; cardiovascular diseases; inflammation

Semiparametric normal transformation models for spatially correlated survival data

Abstract: There is an emerging interest in modeling spatially correlated survival data in biomedical and epidemiologic studies. In this article we propose a new class of semiparametric normal transformation models for right-censored spatially correlated survival data. This class of models assumes that survival outcomes marginally follow a Cox proportional hazard model with unspecified baseline hazard, and their joint distribution is obtained by transforming survival outcomes to normal random variables, whose joint distribution is assumed to be multivariate normal with a spatial correlation structure. A key feature of the class of semiparametric normal transformation models is that it provides a rich class of spatial survival models where regression coefficients have population average interpretation and the spatial dependence of survival times is conveniently modeled using the transformed variables by flexible normal random fields. We study the relationship of the spatial correlation structure of the transformed no...

Piecewise Constant Cross-Ratio Estimation for Association of Age at a Marker Event and Age at Menopause

Abstract: A question of significant interest in female reproductive aging is to identify bleeding criteria for menopausal transition. Although various bleeding criteria, or markers, have been proposed for menopausal transition, their validity has not been adequately examined. The Tremin Trust data were collected from a long-term cohort study that followed a group of women throughout their whole reproductive life. Such data provide a unique opportunity for evaluating the utility of a bleeding criterion-based marker event by assessing the association between age at onset of the bleeding marker and age at onset of menopause. Formal statistical analysis of this dependence is challenged by the facts that both the marker event and menopause are subject to right-censoring and that their association depends on age at the marker event. We propose using the cross-ratio to measure their dependence by assuming the cross-ratio to be a piecewise constant function of age at onset of the marker event. We propose two estimation pro...

Low Bone Mineral Density in the Early Menopausal Transition: Role for Ovulatory Function

Abstract: Objective and Context: The objective of the study was to determine whether luteal abnormalities or measures of sex steroid hormones collected across a menstrual cycle were associated with bone mineral density (BMD) at the total hip or lumbar spine. Design and Setting: The Study of Women’s Health Across the Nation is a longitudinal, community-based study conducted at seven clinical sites. Study of Women’s Health Across the Nation includes a daily hormone study substudy in which daily urine samples are collected for one menstrual cycle (up to a maximum of 50 d) each year. Participants: Participants included 643 pre- and perimenopausal women, aged 43–53 yr. Main Outcome Measures: BMD of the lumbar spine and total hip was measured by dual-energy x-ray densitometry. Daily urine samples were assayed for estrone conjugates, pregnanediol glucuronide, LH, and FSH, and the information from across the menstrual cycle was expressed as area under the curve (AUC). BMD levels were evaluated in relation to three menstrua...

Structural inference in transition measurement error models for longitudinal data.

Abstract: We propose a new class of models, transition measurement error models, to study the effects of covariates and the past responses on the current response in longitudinal studies when one of the covariates is measured with error. We show that the response variable conditional on the error-prone covariate follows a complex transition mixed effects model. The naive model obtained by ignoring the measurement error correctly specifies the transition part of the model, but misspecifies the covariate effect structure and ignores the random effects. We next study the asymptotic bias in naive estimator obtained by ignoring the measurement error for both continuous and discrete outcomes. We show that the naive estimator of the regression coefficient of the error-prone covariate is attenuated, while the naive estimators of the regression coefficients of the past responses are generally inflated. We then develop a structural modeling approach for parameter estimation using the maximum likelihood estimation method. In view of the multidimensional integration required by full maximum likelihood estimation, an EM algorithm is developed to calculate maximum likelihood estimators, in which Monte Carlo simulations are used to evaluate the conditional expectations in the E-step. We evaluate the performance of the proposed method through a simulation study and apply it to a longitudinal social support study for elderly women with heart disease. An additional simulation study shows that the Bayesian information criterion (BIC) performs well in choosing the correct transition orders of the models.